Offline/internal network installs can be handled with flatpak create-usb - https://docs.flatpak.org/en/latest/usb-drives.html

One can distribute flatpaks along with their dependencies on USB drives (or network shares, etc.) which is especially helpful in situations where Internet access is limited or non-existent.

Cache/mirroring would be great for those who need it.

Edit:

Thinking about it, I wonder if there’s enough “core features” with ‘create-usb’ that its just matter of scripting something together to intercept requests, auto-create-usb what’s being requested and then serve the package locally? If a whole mirror is required, it may be possible to iterate over all flathub packages and ‘create-usb’ the entire repo to have a local cache/mirror? Just thinking “out loud”.

Looks like at least one type can take flight from the ground, although with some difficulty: https://www.youtube.com/watch?v=kIl_bYFMr8o

It was introduced two years ago: https://element.io/blog/introducing-native-matrix-voip-with-element-call/

Looks like at least two other clients support 1:1 calls.

For anyone considering Session messenger:

The Session developers dropped Perfect Forward Secrecy because it would be hard to work around it.

First things first, let’s talk about what we’re leaving behind: Perfect Forward Secrecy (PFS) and deniability.

Source: https://getsession.org/session-protocol-explained

In plain English, they dropped a security feature for their convenience to the detriment of their users’ security.

For anyone unsure what PFS provides:

The value of forward secrecy is that it protects past communication.

Source: https://en.wikipedia.org/wiki/Forward_secrecy

The Session devs also claim:

Session provides protections against these types of threats in other ways — through fully anonymous account creation, onion routing, and metadata minimisation, for example.

Reading between the lines, we can interpret that as introducing security through obscurity, which is generally considered bad practice - https://cwe.mitre.org/data/definitions/656.html

Private messages on Matrix have been end-to-end encryption (E2EE) by default since 2020 - https://matrix.org/blog/2020/05/06/cross-signing-and-end-to-end-encryption-by-default-is-here/

For anyone considering Telegram for privacy:

1. Telegram doesn’t default to encryption. All your messages are stored and can be viewed by anyone with enough privileges on Telegram’s infrastructure.
2. Telegram’s “secure” 1-1 messages are limited to the point of being useless and not worth using. It’s a dark design pattern created to discourage their use, ensuring you give them all your data.
3. Telegram doesn’t support E2EE group messages.

TL;DR - Matrix is more private than Telegram.

In Matrix a direct chat is a group chat with two people.

You’re right, I forgot how Matrix handled messages and the current state is that there’s are at least 6 other clients that support E2EE - this is awesome.

That said, as soon as you look for a stable client that supports other features like Native 1:1 calls and Threads the only client listed is Element, check here: https://matrix.org/ecosystem/clients/

Side note: Looks like ~3 years ago a Fluffychat dev stated they would not implement E2EE in the app [0], this must have been around the time I was looking at other clients because I recall this one “looking” the best and might be viable for non-techy people to use/recommend. I’m glad they changed their mind and implemented E2EE. Time to take a look at it again.

[0] https://gitlab.com/KrilleFear/fluffychat/-/issues/25#note_423061121

First off, how can you claim RCS "requires you to buy an Android and then state iMessage is "cross platform through Apple’s ecosystem? RCS works on Android and is available in various devices from many manufacturers. iMessage is only available on devices sold by Apple.

Secondly, why would you rate iMessage higher than RCS for “ease of use”? That makes zero sense, they behave basically the exact same way.

Lastly, RCS is coming to iOS - Apple’s just been lagging because implementing a cross-platform solution is detrimental to their profits.

So RCS will eventually work across iOS and Android AND work by default. There’s no reason RCS wouldn’t be easier or rated higher than iMessage in terms of “ease of use”

This is an often repeated piece of misinformation. The developer of gurk-rs, a third party Signal client, has even said this himself. The client presents itself with a completely identifiable name to the Signal servers - the Signal devs can see this and could easily block this client from connecting but they don’t. This project has existed for at least 3+ years now.

There’s a few clients for Signal, nobody is preventing developers from creating apps; there’s Molly, gurk-rs, Axolotl, Flare, signal-cli, Pidgin (with the Signal plugin.

The problem is 3rd party clients don’t implement all features because it takes a lot of work and they’re created/developed by volunteers - just take a look at Matrix and how many clients support all features or even just group end-to-end encryption (E2EE). Last I checked many third party Matrix clients didn’t support encrypted group messages, primarily just Element, the reference client built by the matrix developers. So you have the same problem on Signal that you have on Matrix.

I could settle for this but remove telegram as it’s not even E2EE by default. It’s basically facebook v2.

Signal > Matrix/Element > RCS > SMS.

iMessage isn’t in the equation because it only works on a single platform.

What’s wrong with Briar? https://briarproject.org/

Censorship-resistant peer-to-peer messaging that bypasses centralized servers. Connect via Bluetooth, Wi-Fi or Tor, with privacy built-in.

I think the reason these apps don’t take off is the compromises they make in order to work the way they do. When you do need them, you best hope you’re able to get them and get others to use them as well.

kinda agree with you.

Firefox is not doing well, and greedy CEO’s are not helping the cause. I wish they’d take a playbook from Nintendo’s leadership, show they really back the product, and take a pay cut to help the cause.

I can’t imaging the CEO being significantly impacted if they had to go from $6,700,000 per year to $3,350,000 and could single-handedly save at least 10 engineers at $300,000+ each to continue to work on core features and guarantee long-term success.

Nintendo’s Satoru Iwata on layoffs:

If we reduce the number of employees for better short-term financial results, employee morale will decrease, and I sincerely doubt employees who fear that they may be laid off will be able to develop software titles that could impress people around the world

out of the loop since I’ve moved to debian and been using flatpak for the last few years, what software are you installing via PPA that isn’t generally available via flatpak?

The flatpak version updates in the background, doesn’t interrupt if its already running, and is immediately on the latest version the next time you run firefox.

It’s disingenuous to act like this is some huge burden.

Having to double your software engineers, UI/UX designers, QA engineers, DevOps, and localization/accessibility specialists to handle a second browser is a HUGE burden for a non-profit.

If you don’t care about quality, security, or user experience, sure you can just pass a “does it compile” test and push to prod. You’ll quickly find that nobody wants to use this under resourced browser.

Or if it’s such a pain, you don’t bother and just ship the WebKit version everywhere.

This is exactly what Apple wants. They don’t want to give people a real choice because they’re scared of real competition.

Ooh silverbullet looks nice too, thanks. Link for the lazy: https://silverbullet.md/

If you’re on Firefox on desktop/laptop, check out Bypass Paywall [0]. It was removed from the firefox add-on store due to a DMCA claim [1], but can be manually installed (and auto updates) from gitlab. The dev even provides instructions on how to add custom filters to uBlock Origin [2], so you don’t have to add another extension but still get some benefit.

[0] https://gitlab.com/magnolia1234/bypass-paywalls-firefox-clean

[1] https://winaero.com/mozilla-has-silently-removed-the-bypass-paywalls-clean-add-on-from-amo/

[2] https://gitlab.com/magnolia1234/bypass-paywalls-clean-filters

CLI’s are likely not specifically the target. I suspect the CLI is just the “low hanging fruit” and core set of software that needs to be supported before you build up to a fully functional GUI apps.

I’ve been hoping this project makes significant progress for the last few years to run GUI apps. unfortunately it’s been slow as there’s not as much interest in getting Mac apps to run on Linux as there is with WINE. that said, I don’t fault them, it’s a daunting task and wine has the benefit of three decades of progress under their belt.